Schizophrenia and other forms of psychosis affect approximately 3% of the population with a disorder that is usually chronic and disabling. The peak age of onset is between ages 18-30, occurring just as life's most productive years are beginning. Although genetic liability and abnormal brain development are known contributing factors, the etiology and pathophysiology of schizophrenia and related syndromes is largely unknown. To date, prospective observation of onset, i.e., the transition from vulnerability to disorder has not been possible because most persons at true risk cannot be identified premorbidly. This has hampered efforts at prevention. However, recent progress in risk ascertainment methodology has enabled reliable identification of help-seeking persons with pre-psychotic or prodromal clinical syndromes who develop psychosis within 1-2 years at rates between 20%-50%. Thus, clinical high-risk populations are now available for tracking prospectively the development and emergence of psychosis. However, because of the low incidence of schizophrenia and the heterogeneity of outcomes in clinical high-risk cases, single site studies cannot efficiently exploit the risk criteria in identifying predictors and mechanisms of psychosis. The NAPLS consortium was created to solve this problem. Eight NIMH-funded sites in North America studying prodromal patients using a common prodromal assessment instrument pooled data to create the largest sample of such persons worldwide (N=291), 35% of whom converted to psychosis after 2 years. An algorithm of baseline data was generated predicting psychosis with about 80% positive predictive power and 40% sensitivity. In this revised proposal, we describe a collaborative prospective study for which we will recruit 800 cases and 400 appropriate controls over 5 years using common, standardized clinical and neurobiological measures. The aim is to collect a sample with sufficient size and power to rigorously test elements critical to the liability for and development of psychosis in the biomarker domains of brain structure, electrophysiology, stress hormones, and genomics, and in the clinical domains of prodromal presentation and epidemiology. The revised proposal addresses reviewers' concerns, including the integration of the research plan and measures into a unifying framework. The findings will enhance our ability to identify persons at high risk for imminent psychosis, by refining predictors of conversion, and expanding our understanding of the underlying neural mechanisms. Such knowledge is critical for future efforts at early detection, intervention and prevention of psychotic disorders.